The organization and the complete nucleotide (nt) sequence of the b5R gene encoding human NADH-cytochrome b5 reductase (b5R; EC 18.104.22.168) have been determined by a combination of restriction mapping and nt sequence analysis of overlapping genomic DNA clones. The entire gene is about 31 kb in length and contains nine exons and eight introns. Exon 2 contains the junction of the membrane-binding domain and the catalytic domain of b5R, indicating that two forms of b5R, a soluble and a membrane-bound form, are generated by post-translational processing. The 5' portion of the b5R gene lacks the canonical 5' transcriptional regulatory elements, but contains five copies of the GC box sequence G-G-G-C-G-G. While the average G + C content of the b5R gene is 55%, that of the 5' portion of the gene is extraordinarily high (86%). The CpG dinucleotide sequence was found at a very high frequency in this G + C-rich region. These structural features are very similar to those of the regulatory regions of constitutively expressed 'housekeeping' genes. Several transcription start points were identified by the primer extension experiment. Seventeen complete and twelve incomplete Alu family sequences were found in introns. An uncanonical polyadenylation signal was detected in the 3'-untranslated region of the gene as A-G-T-A-A-A instead of A-A-T-A-A-A.